Of the 4 million infants born each year in the United States, approximately 60,000 weigh less than 1,500 grams (very low birth weight-VLBW). Over twelve percent of these infants sustain brain injury with subsequent developmental delay. Although various neuroprotective strategies have been evaluated, none have been successful. Effective interventions are desperately needed to treat these most vulnerable of infants. One promising intervention is the use of recombinant erythropoietin (Epo, also known as an erythropoiesis stimulating agent, or ESA). In addition to stimulating red cell production, Epo has been shown to be protective in the developing brain in animal models. We have preliminary data suggesting its efficacy when used in VLBW infants, who are at risk of requiring transfusions, and who are also at risk for brain hemorrhage, hypoxic- ischemic brain injury, and developmental delay. We are currently performing a multicentered study evaluating hematopoietic and short term developmental effects of ESAs in preterm infants randomized to receive Epo, Darbepoetin alfa (a longer acting ESA), or placebo/ control for the first 10 weeks of age. The first enrolled infants will reach 42-48 months in January, 2010. While that study evaluates the safety and general short-term developmental effects of ESAs, there is an unprecedented opportunity to study long term effects of ESA in significant detail, including evaluating the long term developmental effects and the underlying mechanism of neurologic improvement with state of the art multimodal neuroimaging. This proposal seeks to evaluate longitudinal, long-term developmental effects and underlying neurologic mechanisms of ESAs administered to VLBW infants in the first 10 weeks of life. Our specific hypotheses are: 1) ESAs administered to preterm infants during the neonatal period improve long-term neurodevelopmental outcome, 2) ESAs affect regional brain structure, neurochemistry and neurologic organization as reflected in MR imaging, and 3) the blood level of ESA correlates with MR imaging and neurodevelopmental outcome. To test these hypotheses, neurodevelopmental outcome will be assessed through a comprehensive neurodevelopmental assessment at two time points: 42-48 months, and xx-xx months (WPSSI III, Early Child Assessment, Executive Categorization Battery). Brain imaging will be performed concurrent with developmental assessments and includes measures of volume (high resolution volumetric analysis), neurochemistry (magnetic resonance spectroscopy) and regional cerebral blood flow (arterial spin labeling). This study is highly clinically relevant due to the long-term developmental and imaging follow up studies that are part of the design, significantly increasing our ability to determine if developmental, functional and anatomical differences exist in infants randomized to ESAs, a relatively new interventional strategy used in preterm infants. This proposal addresses our long-term goal of developing effective treatment strategies for disorders associated with prematurity through an improved understanding of brain-behavioral relationships.